Bugs item #1520735, was opened at 2006-07-11 12:28
Message generated for change (Comment added) made by sf-robot
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1520735&group_id=56967
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: PF/loader
Group: (zombie: Pathfinder 0.12.0)
>Status: Closed
Resolution: None
Priority: 5
Private: No
Submitted By: Wouter Alink (vzzzbx)
Assigned to: Nobody/Anonymous (nobody)
Summary: XQ: broken dbfarm
Initial Comment:
Sometimes (don't know exactly when it occurs, but at
least Michael Schmidt and I have noticed it) the
dbfarm gets broken when adding documents to the
database. Subsequently shredding new documents
implicitly or explicitly (with the same name) results
in errors like:
ERROR: [rename]: 29 times inserted nil due to errors
at
tuples "pre_size", "pre_level", "pre_prop", "pre_kind"
, "qn_uri", "qn_prefix".
ERROR: [rename]: first error was:
ERROR: rename(<tmp_351>,pre_size2): operation failed
ERROR: interpret_unpin: [rename] bat=171,stamp=-1285
OVERWRITTEN
ERROR: BBPdecref: tmp_253 does not have pointer fixes.
ERROR: interpret_params: +(param 2): evaluation error.
Some of the characteristics:
- so far has been noticed on Linux systems (Suse and
Redhat)
- has been noticed on both 32 and 64 bit machines
- until now has only been noticed with relatively
large documents (100MB+)
- occured in 0.11.3 (few months back) as well as the
0.12.0 release
- occurs with both implicit (using "doc
("/home/.../...xml")") and explicit shredding using
shred_doc("...","...").
- only clearing the dbfarm folder seems to solve the
problem
I can't reproduce it, but if anybody notices the same
error... please let it be known.
p.s. maybe it would be better not to delete the
dbfarm folder when this error occurs, but keep a copy
of it so that it can be analysed ;)
Question for the GDK-gurus among us: could it have
anything to do with the mmap-ing of bats in the
shredder?
----------------------------------------------------------------------
>Comment By: SourceForge Robot (sf-robot)
Date: 2007-12-14 19:20
Message:
Logged In: YES
user_id=1312539
Originator: NO
This Tracker item was closed automatically by the system. It was
previously set to a Pending status, and the original submitter
did not respond within 365 days (the time period specified by
the administrator of this Tracker).
----------------------------------------------------------------------
Comment By: Peter Boncz (boncz)
Date: 2006-12-14 09:31
Message:
Logged In: YES
user_id=591107
Originator: NO
there was a bug in TMsubcommit (gdk_delta.mx) that caused repositories to
corrupt.
(some heaps were not saved)
This has been fixed in the 0.12 => 0.14, september 2006 timeframe
0.12 still has this bug.
In any case, I also find it hard to provide a test script for this one.
----------------------------------------------------------------------
Comment By: Fabian (mr-meltdown)
Date: 2006-10-26 07:03
Message:
Logged In: YES
user_id=963970
noticed anything lately?
----------------------------------------------------------------------
Comment By: Wouter Alink (vzzzbx)
Date: 2006-07-13 14:00
Message:
Logged In: YES
user_id=621590
sorry, my fault. in my last comment i meant: 0.12.1
----------------------------------------------------------------------
Comment By: Peter Boncz (boncz)
Date: 2006-07-13 07:43
Message:
Logged In: YES
user_id=591107
*please* be explicit whether you are working on the HEAD or
in the Stablee
----------------------------------------------------------------------
Comment By: Wouter Alink (vzzzbx)
Date: 2006-07-11 12:45
Message:
Logged In: YES
user_id=621590
had a look at shredder.mx (CVS-HEAD): from line 602 and
1644 it appears that no post processing is done when the
libxml halts. So my question: what happens with the non-
finalized (possibly mmap-ed) bats (dstBAT[]) in that case?
Shouldn't there be some cleaning-up code?
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1520735&group_id=56967
-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Monetdb-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs
